Differential effect of Bacillus firmus on immune response and enterocyte brush-border enzyme levels in BALB/c and B10.BR mice

A nonpathogenic bacterium of external environment possessing remarkable immunomodulatory activity, Bacillus firmus (BF) inactivated with formaldehyde, was given intragastrically to two genetically different mouse strains BALB/c (H-2d) and B10.BR/SnPh (B10.BR, H-2k) reared in conventional (CV) and B10.BR strain also in germ-free (GF) conditions. Repeated intragastric administration of BF (500 micrograms every other day over two weeks, starting at the age of 3 months) significantly enhanced intestinal IgA levels in CV BALB/c mice but did not affect intestinal IgA in CV B10.BR mice. In GF B10.BR mice, IgG levels in sera and intestinal washings increased after BF administration compared to CV B10.BR mice. In CV BALB/c mice, specific activity of enterocyte brush-border enzymes (lactase, gamma-glutamyltransferase, alkaline phosphatase) decreased after BF treatment; sucrase (sucrose alpha-glucosidase) activity was not affected. On the other hand, in B10.BR mice, specific activity of gamma-glutamyltransferase and dipeptidyl peptidase IV were higher after administration of BF in both CV and GF groups relative to untreated controls. The activities of lactase and glucoamylase (glucan 1,4-alpha-glucosidase) were significantly stimulated only in the group of GF B10.BR mice treated with formolized BF. The stimulation of immunoglobulin production after BF treatment was accompanied by changes in the levels of enterocyte brush-border enzymes; this responsiveness to BF treatment was genetically regulated.     

Evidence for clustering of H-2K, H-2D, and the Fc receptor on the membranes of B cells.

Alloantisera to H-2K, H-2D, and Ia antigens markedly inhibited the binding of EA but not FITC-IgG by the B cell Fc receptor. EA rosette formation approached normal levels when masked H-2 but not Ia specificities were allowed to cap on the membranes of B cells. beta2-mu coated SRBC were bound by the Fc receptor, and high concentrations of soluble beta2-mu were found to moderately inhibit EA rosette formation while lower concentrations enhanced binding. The data support the concept of Fc/Ia identity, and they suggest that H-2K, H-2D, and the Fc receptor may be closely grouped on the membranes of B cells. Further, these observations suggest that the beta2-microglobulin associated with H-2 could serve to link T cells with the Fc receptor of B cells during the inductive phase of antibody synthesis.     

Demonstration of Ia antigens on mouse intestinal epithelial cells by immunoferritin labeling

Several kinds of epithelial cells that express H-2 antigens were studied by immunoferritin labeling with an antiserum reacting only with antigens of theI region of theH-2 complex. Spleen lymphocytes were used to test the labeling system and the effect of the epithelial cell dissociation procedure on Ia antigens. Immunoglobulin-positive B10.BR lymphocytes were labeled with an anti-la^k serum (A.TH anti-A.TL serum absorbed with BALB/c and B10.D2 cells), while congenic B10.D2 lymphocytes were unlabeled. The distribution of labeled Ia antigens on living B10.BR lymphocytes was patchy, while on cells fixed in periodate-lysine-paraformaldehyde before labeling, the distribution of label was continuous. Fixation evidently immobilized Ia antigens in the lymphocyte membrane. Trypsin and collagenase, as used in the epithelial cell dissociation procedure, had no discernible effect on the Ia antigens of lymphocytes. The epithelial cells studied included the columnar absorptive cells of the small intestine, uterine lining epithelium, tracheal brush cells, and pancreatic exocrine and duct cells. These cells were fixed before dissociation from their respective tissues. Ia antigens were detected only on the columnar absorptive cells of the small intestine. These cells labeled equally well with an antiserum reacting only with theK -end of theH-2 complex. In both cases, congenic control intestinal cells were unlabeled. Thus, intestinal epithelial cells appear to express theIa, K, and presumablyD regions of theH-2 complex, while the other epithelial cell types express only the K and D antigens. On fixed intestinal epithelial cells, Ia and H-2K antigens were continuously distributed on the lateral and basal cell membranes including the zonula adherens, but the antigens were absent from the apical microvillous membrane and the zonula occludens.     

Structural studies of the protein portion of the H-2-linked Ia glycoprotein antigens of the mouse: tryptic peptide comparison of products from the I-A and I-C subregions of B10-HTT.

Ia antigens from the I-A8 and I-Ck subregions of the B10.HTT (H-2t3) strain of mice were isolated by indirect immunoprecipitation of arginine-labeled, nonionic detergent-solubilized materials. After biochemical purification the electrophoretically homogeneous 28,000 dalton glycoprotein beta chains from the Ia precipitates were digested with trypsin and the resultant radiolabeled tryptic peptides were compared by analytical ion exchange chromatography. These comparisons reveal that the beta chains of Ia antigens from the A (I-A8) and C (I-Ck) subregions of B10.HTT share only two out of 12 to 14 of their arginine tryptic peptides. Thus these noncross-reactive Ia antigens are structurally quite diverse, and would possess sufficient structural variability to account for their lack of antigenic cross-reactivity.     

Proliferating and helper T lymphocytes display distinct fine specificities in response to human fibrinopeptide B

The fine specificity of T cell responses involved in the generation of help for antibody production and proliferation was examined by using the 14 amino acid peptide human fibrinopeptide B (hFPB, B beta 1-14) and its synthetic peptide homologues B beta 1-14(Lys14), B beta 1-13, and B beta 3-14. Peritoneal exudate or lymph node T cells from C57BL/10 and B10.BR mice immunized with hFPB or its synthetic homologues were used to measure in vitro proliferative responses. T cells from hFPB-immunized B10.BR mice showed specific proliferation to hFPB, but were unresponsive to B beta 1-14(Lys14), B beta 1-13, and B beta 3-14. B10.BR mice immunized with B beta 1-14(Lys14), B beta 1-13, or B beta 3-14 were unresponsive to all peptides tested. T cells from C57BL/10 mice showed no specific proliferation after immunization and challenge with any of the peptide antigens. In contrast to the patterns of T cell proliferation, immunization of both B10.BR and C57BL/10 mice with hFPB, B beta 1-14(Lys14), B beta 1-13, or B beta 3-14 primed for significant helper T cell activity, as assessed by the augmentation of a primary in vitro B cell IgM anti-FITC plaque-forming cell response after culture with B beta 1-1(Lys14)-FITC. Significant peptide-specific helper activity was observed when the FITC moiety was conjugated to the carboxyl terminal lysine (B beta 1-14(Lys14)-FITC) as well as FITC substitution at the amino terminus (FITC-B beta 3-13 or FITC-B beta 3-14). These results suggest that the fine specificity of T cell responses to peptide antigens are different for helper and proliferating T cells and that responsiveness by one T cell subpopulation does not predict the response pattern of other functional subpopulations.     

Neonatal tolerance induction in the thymus to MHC-class II-associated antigens. I. Preferential induction of tolerance to Mls antigens and resistance to allo-MHC antigens

Neonatal tolerance inducibility of self-major histocompatibility complex (MHC)-class II-associated antigens was compared with that of allo-class II antigens. BALB/c (H-2d, Mlsb) mice, less than 24 hr after birth, were intravenously injected with bone marrow cells of either (BALB/c X DBA/2)F1 (H-2d, Mlsb/a, semiallogeneic at the Mls locus) or (BALB/c X B10.BR)F1 (H-2d/k, Mlsb; semiallogeneic at the MHC), as antigens. The mice were tested for in vivo immune activity of class II-reactive T cells by means of the popliteal lymph node-swelling assay. They developed tolerance, irrespective of type of antigens, showing profoundly suppressed host-versus-graft reaction, and those tolerized to the allo-MHC antigens accepted skin grafts of the corresponding allogeneic mice. In the thymus and spleen of the Mls-tolerant mice, antigen-specific class II-reactive T-cell activity was completely abolished, without the apparent involvement of suppressor cells. In contrast, the activity in allo-MHC-tolerant mice was not reduced in either thymus or peripheral lymphoid organs, suggesting that systemic hyporesponsiveness is attributable to reversible suppression of immune competent cells. The resistance for cell-level tolerance induction to allo-class II antigens may not be ascribed to the active participation of allo-MHC antigens in prevention of or in escape from tolerance induction or both, since an injection of bone marrow cells of both Mls and H-2-semiallogeneic (DBA/2 X B10.BR)F1 (H-2d/k, Mlsa/b) mice could induce tolerance to Mlsa-H-2d antigens in newborn thymus cells.     

Detection of at least two distinct mouse I-E antigen molecules by the use of a monoclonal antibody

In an attempt to determine whether the expression of more than a single Ia antigen is determined by the I-E subregion of the mouse major histocompatibility complex (MHC), sequential immunoprecipitation analyses were performed by using a monoclonal antibody and alloantisera reactive with I-E subregion products. 3H-leucine-labeled glycoprotein preparations obtained from H-2d spleen cells were precleared with the monoclonal antibody 14-4-4S and then examined for residual Ia activity precipitable by an alloantiserum detected by SDS-polyacrylamide gel electrophoresis. Residual Ia activity was observed for all three strains of the H-2d haplotype tested. The residual Ia activity could be detected by sera absorbed with B10.A spleen cells, indicating that products of the I-E subregion rather than of the I-C subregion were responsible for this activity. No separable I-Ek molecules were detected in products of B10.A cells with the use of combinations of two monoclonal antibodies (including 14-4-4S) and several appropriate alloantisera. These findings indicate the presence of at least two similar but distinct I-E antigens encoded by the H-2d haplotype.     

Ia specificities on parental and hybrid cells of an I-A mutant mouse strain

Splenic B cells and B cell blasts from the I-A mutant mouse strain B6.C-H-2bm12 were tested by serology with a series of new monoclonal anti-Iab antibodies. Four out of 5 of those monoclonal antibody-defined specificities that are determined by wild-type I-Ab antigens were undetectable on B6.C-H-2bm12 cells. Specificities both present and absent on mutant cells appear to be determinants on the same wild-type molecule, as indicated by sequential precipitation experiments with soluble H-2b antigens. The lack of expression of certain Ia specificities on mutant cells was found not to be the result of disparate control by the Xid gene, which was previously shown to control the expression of Ia.W39, another specificity absent in B6.C-H-2bm12 mice. Serologic testing of Ia specificities on cells and blasts from F1-hybrid mice suggested that the Iabm12 antigens are codominantly expressed, indicating a failure to detect trans regulation or complementation of the mutant phenotype. Another monoclonal antibody-defined Ia specificity dependent on the expression of the E beta polypeptide was normally expressed in B6.C-H-2bm12 mice. These data thus suggest that the lesion of these mutant mice occurred in the A alpha and/or A beta structural gene, resulting in the loss of several Ia specificities.     

Ia Restriction Specificity of KLH-Specific T Cells from Allogeneic Bone Marrow Chimeras Is Influenced by Histocompatibility at the H-2 and Minor Histocompatibility Loci

Ia restriction specificity involved in T cell proliferative responses to keyhole limpet hemocyanin (KLH) has been analyzed using a variety of allogeneic bone marrow chimeras. The chimeric mice were prepared by reconstituting irradiated AKR, SJL, B10.BR and B10.A(4R) mice with bone marrow cells from B10 mice. When such chimeric mice had first been primed with KLH in complete Freund's adjuvant (CFA), T cells from H-2 incompatible fully allogeneic chimeras showed significantly higher responses to KLH in the presence of antigen-presenting cells (APC) of donor strain (B10) than APC of recipient strain. However, in H-2 subregion compatible chimeras, [B10→B10.A(4R)], which were matched at the H-2D locus and at minor histocompatible loci, the T cells could mount vigorous responses to KLH with antigen-presenting cells (APC) of either donor or recipient type. The same results were obtained as well with chimeras that had been thymectomized after full reconstitution of lymphoid tissues by donor-derived cells. A considerable proportion of KLH-specific T cell hybridomas established from [B10→B10.A(4R)] chimeras exhibited both I-A^b and I-A^k restriction specificities. The present findings indicate that the bias to donor Ia type of antigen specific T cells is determined by donor-derived APC present in the extrathymic environment but that cross-reactivity to the recipient Ia is influenced to some degree by histocompatibility between donor and recipient mice, even though the histocompatible H-2D locus and minor histocompatibility loci seem not to be directly involved in the I-A restricted responses studied herein.     

Sharing of Ia antigens between species. II. Molecular localization of shared Ia determinants implies the existence of more than one I sublocus of the rat MHC.

A mouse alloantiserum B10.D2 anti-B10.BR (H-2d anti-H-2k) cross-reacted with rat lymphocyte surface glycoproteins with characteristics of Ia antigens. Sequential precipitation analysis on solubilized radiolabeled LEW rat lymphocyte antigens with this cross-reactive mouse alloantiserum and the rat alloantiserum BN anti-LEW (Ag-B3 anti-Ag-B1) revealed that the Ia-like antigen detected by the mouse alloantiserum also reacted with the rat anti-Ia antibodies. It was further shown that the rat alloantiserum also detected another set of Ia-like antigens that did not cross-react with the mouse alloantibody. Precipitation analysis with congenic rat strains confirmed that all Ia-like antigens precipitated by the rat alloantibody were encoded by Ag-B linked genes. Thus the shared Ia-like antigen must also be the product of Ag-B-linked gene(s) or be physically associated with such products. In addition, molecules bearing shared antigenic determinants were separable from at least some of the Ia-like antigens detected by the rat alloantiserum, possibly suggesting the existence of more than one sublocus coding for Ia antigens within the rat MHC.     

Levels of Endogenous lnterleukin-1, Interleukin-6, and Tumor Necrosis Factor in Congenic Mice Infected with Borrelia garinii

This study describes the levels of interleukin-1 alpha (IL-1α), tumor necrosis factor alpha (TNFα) and interleukin-6 (IL-6) in the sera and parenchymal organs of various congenic mouse strains infected with Borrelia garinii. A significant elevation of inflammatory cytokine levels was found in the organs of C3H/HeN (H-2^k) and B10.BR (H-2^k) mice but not in those of BALB/c mice (H-2^d). Focally produced cytokines can contribute to antimicrobial defense against these organisms. High levels of IL-1α were observed in the sera of C3H/HeN, B10.BR and B10 (H-2^b) mice infected with B. garinii and they were associated with the presence of spirochetes in the skin. Thus, susceptible mice demonstrated a stronger cytokine response than resistant mice. This study presents in vivo evidence that B. garinii infection affects the immunopathogenesis of Lyme disease.     

The distribution of Ia antigens on the surfaces of lymphocytes.

The distribution of Ia antigens was studied on murine spleen lymphocytes by an ultrastructural technique employing deep freeze-etched replicas. Ia antigens were labeled on cells from appropriate congenic and recombinant strains of mice by incubating the cells with FITC-conjugated anti-Iak antibody, followed by ferritin-coupled Fab anti-FITC. Ia antigens were detected predominantly on immunoglobulin (Ig)-bearing B lymphocytes. Antigens coded for by the entire Ik region were present on the surfaces of 95% of the positive cells (from B10.BR mice) in densely packed microclusters. Ia specificities coded for by the I-A and I-C subregions (on 4R and B10.HTT mice) exhibited a more variable pattern, with 30 to 35% of the labeled cells having sparsely distributed Ia antigens in relatively discrete microclusters. Binding of anti-Iak antibody at 37 degrees C led to patch formation but not to capping. Modulation of surface Ig left Ia antigens diffusely distributed on the cell surface, indicating that these two membrane proteins are independent molecules.     

Heterogeneity in the immune response to serotype b LPS of Actinobacillus actinomycetemcomitans in inbred strains of mice

We investigated the heterogeneity of the humoral immune responses to whole cells and lipopolysaccharide (LPS) of Actinobacillus actinomycetemcomitans serotype b and production of cytokines in inbred strains of mice. Nine such strains were tested: A/J (H-2(a)), C57BL/6 (H-2(b)), BALB/c (H-2(d)), DBA/2 (H-2(d)), B10.BR (H-2(k)), C3H/He (H-2(k)), C3H/HeJ (H-2(k)), DBA/1 (H-2(q)) and B10.S (H-2(s)). Mice were immunized intraperitoneally with whole cells of A. actinomycetemcomitans ATCC 43718 (serotype b) in phosphate buffered saline (PBS; pH 7.2) emulsified with an equal volume of Freund's incomplete adjuvant. Serum immunoglobulin G (IgG), immunoglobulin A (IgA) and immunoglobulin M (IgM) levels against A. actinomycetemcomitans were measured by an ELISA system. ELISA analysis, using LPS fractions from serotype a, b or c strains of A. actinomycetemcomitans as the coating antigens, revealed that mice strains C3H/He, C3H/HeJ, B10.BR and B10.S had an extremely high-IgM response against serotype b LPS. High-IgM titer sera contain also elevated levels of IgA antibodies to the antigen. To compare the cytokine production among inbred mice, the amounts of interleukin-4 (IL-4), interleukin-5 (IL-5), and interleukin-6 (IL-6) released from mouse splenocytes were measured using ELISA systems specific for these cytokines. A. actinomycetemcomitans serotype b LPS stimulation induced IL-6 release from murine splenocytes of all tested strains. However, IL-4 and IL-5 were detected only in high-IgM/IgA responders to A. actinomycetemcomitans serotype b LPS, not in low-IgM/IgA responders. Thus, we found a relationship between the humoral immune response to LPS of A. actinomycetemcomitans serotype b and production of type 2 cytokines by splenocytes.     

Separation of the immune response genes for LDH-B and MOPC-173. I. Description of an immune response defect in B10.BASR1

By using the intra-I-region recombinant mouse strain, B10.BASR1 (H-2as4), the immune response (Ir) genes for LDH-B and MOPC-173 were genetically and serologically separated, as assayed by T cell proliferation. Previous work demonstrated that the H-2s and H-2b strains respond to LDH-B and MOPC-173, whereas the H-2a and H-2k strains failed to respond due to haplotype-specific suppression of I-Ak-activated T helper cells by I-Ek-activated T suppressor cells. In the experiments reported here, B10.BASR1 mice, which lack I-Ek expression, mounted a significant T cell proliferative response to MOPC-173 but not to LDH-B. Separation of the Ia determinants used in restricting these two antigen responses was further confirmed when pretreatment of B10.S(9R) (A alpha sA beta sE beta sJk) macrophages with A.TL anti-B10.HTT (anti-A beta sE beta sJs) serum absorbed with B10.BASR1 spleen cells blocked the LDH-B response but not the MOPC-173 response. Unabsorbed serum blocked both antigen responses. The primary immunogenic determinant recognized by LDH-B or MOPC-173 immune T cells was not present on both antigens, as MOPC-173-primed T cells and LDH-B-primed T cells responded only to the priming antigen. Lastly, by using the A beta mutant strain, B6CH-2bm12, it was shown that the Ir gene and Ia determinants affected by this mutation had no effect on the LDH-B and MOPC-173 proliferative responses. These results suggest the possibility of an intragenic recombinatorial event in either the A alpha or A beta chain resulting in the separation of these two immune response gene functions.     

The influence of self-MHC and non-MHC antigens on the selection of an antigen-specific T cell receptor repertoire

We have examined the influence of self-Ag on TCR expression and specificity in the immune response to the Ag pigeon cytochrome c. Previous work has shown that most Ek-restricted cytochrome c-specific T cells from B10 background mice express TCR alpha beta-heterodimers encoded by V beta 3 and V alpha 11 genes, but that T cells expressing V beta 3 proteins are eliminated due to self-tolerance in Mls-2a mouse strains. Thus, EK-restricted cytochrome c-specific T cells from Mls-2a mice fail to express any V beta 3. In the current study the influence of self-MHC and non-MHC Ag on TCR usage in the immune response to cytochrome c was further examined. First, it was demonstrated that the absence of V beta 3 expression in Mls-2a mice does not alter Ir gene function. Specifically, Mls-2a/Eb haplotype V beta 3- [C3H.SW x B10.A(5R)]F1 mice were high responders to cytochrome c despite the fact that previous structure function analyses have shown a very close correlation between Eb-restricted cytochrome c recognition and V beta 3 expression. This demonstration of the plasticity of TCR expression suggests that relatively few Ir gene defects result from tolerance induced by self-Ag. We also examined differences in V alpha 11 expression among cytochrome c-specific T cells from various H-2k haplotype mouse strains. In particular, the low level of expression of V alpha 11 in cytochrome c-specific T cells from C57BR (H-2k) mice was shown not to be due to self-tolerance. Rather, evidence for limited strain polymorphism of V alpha 11 genes, plus the fact that cytochrome c-specific T cells from F1 hybrids between H-2k, Mls-2b identical C57BR and B10.BR mice express high levels of V alpha 11, suggested the possibility that the variable V alpha 11 usage in the cytochrome c-specific responses of these two strains reflected differences in positive selection during ontogeny by non-MHC non-Mls self-Ag.     

Effects of masking H-2 and Ir gene products on the in vitro responses of cells from immunized and naive mice.

Mouse spleen cells from immunized and naive animals were cultured for 3 days with SRBC and antisera to H-2K, H-2D, Ia, or aggregated IgG. Direct and indirect antibody responses were impaired by each of these agents only when the responding cells had been obtained from unimmunized mice. It was found that as lymphoid cells reacted to the antigen and progressed toward a fully matured antibody response, they first became resistant to inhibition by anti-H-2K and H-2D sera and somewhat later they lost their vulnerability to agents that reacted with Ia antigens.     

Partial characterization of Ia antigens from murine lymphoid cells

Congenic anti-Ia antisera were used to bind radiolabelled Ia antigens from cells of various strains of mice of knownH-2 haplotype. The results indicate that Ia antigens are proteins of molecular weight 30,000 to 35,000 daltons. The Ia antigens are distinct from known H-2 antigens as judged by independent immunoprecipitation as well as by molecular weight. Ia antigens are synthesized by, and are present on the surface of lymphoid cells as evidenced by incorporation studies using³H-leucine and enzymatic radioiodination of cells, respectively. Tissue distribution of cell surface Ia suggests that Ia antigens are on B cells. Ia antigens were detected in the incubation media of³H-leucine labeled splenocytes suggesting that antigens may be secreted.     

Polyclonal B cell activation by B cell differentiation factor B151-TRF2. I. Involvement of self-Ia recognition process mediated by B cells

The present study examined the functional role of Ia antigens on B cells in polyclonal B cell activation induced by a B cell differentiation factor, B151-TRF2. The polyclonal IgM PFC responses by B151-TRF2 were inhibited by monoclonal antibodies specific for class II MHC antigens (Ia antigens) but not class I MHC antigens. Such inhibition by anti-Ia antibodies was haplotype-specific and was observed in the absence of both T cells and accessory cells. Moreover, the anti-Ia antibody-induced inhibition of the B151-TRF2 responses was not due to the blocking of binding of B151-TRF2 to the corresponding B cell receptor. A series of kinetic studies revealed that some Ia-mediated cellular activation process occurs before the resting B cells become responsive to B151-TRF2. Thus, the B151-TRF2-mediated B cell responses consist of at least two distinct phases. The early phase is an Ia-dependent but B151-TRF2-independent process, whereas the late phase is an Ia-independent but B151-TRF2-dependent process. To further characterize the functional role of Ia antigens on B cells, an additional experiment was carried out by using F1 B cells which co-dominantly express both parental Ia antigens on the surface. Interestingly, it was observed that the degree of inhibition of the B151-TRF2-mediated responses of F1 B cells by anti-parental Ia antibody was, at best, one-half that of the parental B cells, suggesting that F1 B cells may be separated into two subpopulations with the restriction specificity for the respective parental Ia antigens. To examine this possibility, (B10 X B10.BR)F1 B cells were separated into adherent and nonadherent cell populations by their ability to bind to either one of the parental B cell monolayers, and the specificity of inhibition of their responses to B151-TRF2 by anti-Ia antibodies was assessed. It was found that the responses of (B10 X B10.BR)F1 B cells adherent to the B10 B cell monolayer or the B10.BR B cell monolayer were almost completely inhibited by anti-I-Ab and anti-I-Ak antibodies, whereas those of nonadherent cells were now selectively inhibited by anti-I-Ak and anti-I-Ab antibodies, respectively. These findings are interpreted as indicating that the B151-TRF2-responsive F1 B cells consist of at least two subpopulations with the restriction specificity for either one of the parental Ia antigens.(ABSTRACT TRUNCATED AT 400 WORDS)     

Sex-dependent effects of non-H-2 loci on lethal GVH reaction induced across an H-2 barrier

We have studied the influence of DBA/2 non-H-2 antigens on the lethal graft-versus-host reaction (GVHR) developed across an H-2 barrier. (DBA/2 x B10.D2)F₁ x B10.D2 (H-2 ^d) backcross (BC) mice were typed for their allelic constitution at nine genetically independent chromosome markers and used as individual cell donors simultaneously for two to three (DBA/2 X B10.D2)F₁ recipients incompatible for DBA/2 non-H-2 antigens alone and two to three (DBA/2 x B10.BR)F₁ recipients incompatible for DBA/2 non-H-2 antigens and H-2^k. The results showed that, when compared with that developed in a control group incompatible for H-2 ^kalone [B10.D2(B10.D2xB10.BR)F₁], the GVHR mortality seen in the presence of an additional incompatibility for DBA/2 non-H-2 antigens [(DBA/2 X B10.BR)F₁recipients] is significantly delayed but only in female mice. An analysis of individual BC donors indicated that this protective effect of DBA/2 non-H-2 antigens correlates with incompatibility for gene(s) linked to the Pgm-1 chromosome marker. In contrast, incompatibility for gene(s) linked to Mod-1 and Es-3 markers accelerates GVHR mortality, but only in male mice. Finally, the results obtained with (DBA/2 x B10.D2)F₁ and (DBA/2 x B10.BR)F₁ recipients were compared; they showed that the intensity of the GVHR developed by cells from individual BC donors against a given set of DBA/2 non-H-2 antigens correlates well with that developed by the same BC donor against the same set of non-H-2 antigens plus H-2^k. We conclude that certain non-H-2 genes (and antigens) can modulate the intensity of the GVHR developed across an H-2 barrier. The number of such genes is probably great; their effects are strong and complex, and can be sex-dependent.